Intelligent bench blasting method and system for rock-mud staggered surface mine

Abstract

The invention discloses an intelligent bench blasting method and system for a rock-mud staggered surface mine, and relates to the technical field of blasting, the method comprises the following steps: acquiring point cloud data of a bench blasting area of the surface mine through three-dimensional laser scanning; constructing a three-dimensional geologic model reflecting the step form and the rock mud distribution characteristics; based on bench blasting characteristics, a comprehensive blasting prediction model is established; constructing a multi-objective optimization function including flying stone control, vibration control and lumpiness uniformity; processing the multi-objective optimization function through a multi-objective optimization algorithm to determine reference hole net parameters, single-hole explosive quantity and millisecond delay of bench blasting; and according to the local rock mud distribution, carrying out self-adaptive adjustment on the reference hole net parameters, the single-hole dosage and the differential delay. According to the method, through three-dimensional geological modeling, multi-target parameter optimization and blasting parameter self-adaptive adjustment, the blasting problem under the rock-mud staggered geological condition is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of blasting, in particular to an intelligent bench blasting method and system for rock and mud staggered open-pit mines. BACKGROUND

[0002] Open-pit bench blasting is a stone blasting operation in the form of benches on the ground. As the main means of mining and stripping in open-pit mines, open-pit bench blasting has made great technological breakthroughs with the rapid development of blasting equipment and rock drilling equipment. The mining scale of mines is continuously expanding, creating significant economic and social benefits.

[0003] However, due to the changing geological conditions of mine engineering, such as the complex interface formed by the staggered distribution of rock and mud, the reflection, refraction and energy attenuation of blasting stress waves will occur when they encounter the mudstone interface during propagation, resulting in uncontrollable distribution of blasting energy and uneven blockiness after blasting. At the same time, the blasting point is close to residential buildings, making blasting vibration control a key limiting factor, and traditional empirical methods are difficult to cope with.

[0004] A Chinese patent with publication number CN117781796A discloses a precise blasting construction method for mudified interlayer surrounding rock tunnel, which comprises: designing blasting holes according to the tunnel excavation section, the holes include undercutting eyes, auxiliary eyes and peripheral eyes arranged in turn from the center of the excavation section to the excavation contour line, the undercutting eyes are provided with a shaped charge cavity structure, and the peripheral eyes include spaced charging holes and empty holes; according to the designed holes, measure the line, lay the hole position and drill the hole; determine the charging parameters and charging structure of the holes; connect the hole charging and the initiation network; check and analyze the blasting results after initiation, and adjust the blasting parameters as needed. This method introduces a shaped charge cavity structure through undercutting eyes and arranges charging holes and empty holes in peripheral eyes at intervals, and adjusts the peripheral eye blasting parameters based on the different positions of the mudified interlayer relative to the tunnel excavation contour line, which can reduce the overbreak and underbreak rates and safety risks caused by the mudified interlayer, improve the blasting construction efficiency, and has wide popularization significance.

[0005] A tunnel blasting vibration prediction method suitable for sand and mud interbedded area is disclosed in Chinese Patent Publication No. CN120627820A, mainly relates to the field of tunnel blasting technology; including steps: S1, determining the key factors affecting the vibration velocity; S2, establishing a sand and mud interbedded area blasting vibration prediction model; S3, data normalization processing; S4, introducing BP neural network to train the prediction model; S5, introducing particle swarm optimization algorithm and initializing particle swarm related parameters; S6, improving the particle swarm optimization algorithm; S7, obtaining the sand and mud interbedded tunnel blasting vibration prediction model of the improved particle swarm optimization BP neural network, and using the model combined with the blasting characteristic parameters of the specific sand and mud interbedded area tunnel engineering, to predict and obtain the blasting effect prediction result; The invention can provide important theoretical guidance for blasting vibration prediction of rock layer variable layered structure, and provide a practical safety evaluation tool for ground vibration velocity control of similar engineering. SUMMARY

[0006] The technical problem to be solved by the present application is to provide an intelligent bench blasting method and system for rock and mud staggered open-pit mines to address the shortcomings of the prior art.

[0007] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is:

[0008] The intelligent bench blasting method for rock and mud staggered open-pit mines comprises the following steps:

[0009] Step S1, obtaining bench blasting area point cloud data of the open-pit mine through three-dimensional laser scanning;

[0010] Step S2, constructing a three-dimensional geological model reflecting the bench shape and rock and mud distribution characteristics;

[0011] Step S3, constructing a comprehensive blasting prediction model based on the bench blasting characteristics;

[0012] Step S4, constructing a multi-objective optimization function including fly rock control, vibration control and block uniformity;

[0013] Step S5, determining the bench blasting reference hole pattern parameters, single-hole charge and millisecond delay by processing the multi-objective optimization function through a multi-objective optimization algorithm, wherein the hole pattern parameters include hole spacing, row spacing and overburden;

[0014] Step S6, adaptively adjusting the reference hole pattern parameters, single-hole charge and millisecond delay according to the local rock and mud distribution.

[0015] Further, the step S2 specifically comprises the following steps:

[0016] Step S2.1, preprocessing the obtained point cloud data through point cloud registration, and removing noise by using a statistical filtering algorithm;

[0017] Step S2.2, extracting step geometry features based on the denoised point cloud data, including step height, slope angle and platform width;

[0018] Step S2.3, identifying the mud boundary based on the color feature, curvature feature and spatial distribution feature;

[0019] Step S2.4, generating a three-dimensional geological model by a Delaunay triangulation algorithm, integrating the mud classification result and the location information of the civilian house.

[0020] Further, the step S2.3 specifically includes the following steps:

[0021] Respectively constructing the color feature component, the curvature feature component and the spatial distribution feature component of each point in the point cloud;

[0022] Comprehensively obtaining the mud classification comprehensive feature value of each point, i.e. the probability of the point belonging to rock or mud, from the feature components of each point;

[0023] The calculation formula of the color feature component is:

[0024]

[0025] Wherein, represents the color feature component of point , represents the point index, respectively represents the red channel intensity value, the green channel intensity value and the blue channel intensity value of point , represents the rock color range set, the RGB value (120, 120, 120) to (200, 200, 200), represents the mud color range set, the RGB value (80, 60, 40) to (150, 120, 80); The calculation formula of the curvature feature component is:

[0026]

[0027]

[0028] Wherein, represents the curvature feature component of point , represents the gradient operator, represents the unit normal vector at point , represents the absolute value operator; The calculation formula of the spatial distribution feature component is:

[0029]

[0030] ​​​​

[0031] wherein, denotes the spatial distribution feature component of a point denotes the total number of points in the neighborhood of a point denotes the jth point in the neighborhood of a point denotes the indicator function, which takes 1 if the condition is true, otherwise takes 0, denotes the set of points belonging to the same material region as a point

[0032] Further, in the step S3, the comprehensive blasting prediction model includes three models, which are: a blasting vibration prediction model considering the elevation effect and rock mass damage; a flyrock distance prediction model integrating surface flyrock, blocked section flyrock and geological defect flyrock; and a Kuz-Ram model oriented to block size prediction.

[0033] Further, in the step S4, the specific formula of the multi-objective optimization function is:

[0034]

[0035] wherein, denotes the multi-objective optimization function vector composed of three objective function values, denotes the vibration control objective function, denotes the flyrock control objective function, denotes the block size uniformity objective function, denotes the blasting parameter vector.

[0036] Further, in the step S5, the multi-objective optimization algorithm adopts a non-dominated sorting genetic algorithm, which includes the following specific steps:

[0037] Step S5.1, randomly generate an initial population, each individual in the population is a solution vector containing hole pattern parameters, single-hole charge and millisecond delay;

[0038] Step S5.2, for each individual in the population, calculate the three objective function values in the multi-objective optimization function respectively;

[0039] Step S5.3, non-dominant sort the population according to the objective function values;

[0040] Step S5.4, calculate the crowding degree of each solution;

[0041] Step S5.5, use binary tournament selection to select parent individuals, and then perform crossover and mutation operations on the selected parent individuals to generate a child population; ​​​​

[0042] Step S5.6, merging the parent population and the offspring population into a new population;

[0043] Step S5.7, performing non-dominated sorting and crowdedness calculation on the new population again, and finally selecting the next generation population;

[0044] Step S5.8, stopping outputting the Pareto optimal solution set when the maximum iteration number 200 is reached;

[0045] Step S5.9, selecting the optimal solution from the Pareto optimal solution set, that is, the reference hole pattern parameters, single-hole charge and millisecond delay.

[0046] Further, in the step S6, the hole pattern adaptive adjustment specifically includes hole spacing adaptive adjustment, row spacing adaptive adjustment and super depth adaptive adjustment;

[0047] The hole spacing adaptive adjustment considers the reference hole spacing, local rock and mud distribution ratio and lithology adjustment coefficient, and scales the reference hole spacing according to the local rock rate and mud rate;

[0048] The row spacing adaptive adjustment establishes an association relationship with the hole spacing through the row spacing coefficient based on the adjusted hole spacing and step height factor, and considers the correction when the step height deviates from the standard value;

[0049] The super depth adaptive adjustment considers the reference super depth, local lithology difference and step height influence, adjusts the super depth value according to the proportion difference of rock and mud, and compensates for the non-standard step height.

[0050] Further, in the step S6, the single-hole charge adaptive adjustment is based on the reference single-hole charge obtained by optimization, and dynamically adjusts in combination with the local lithological characteristics, the relative distance between the blast hole and the civilian house and the vibration control requirement.

[0051] Further, in the step S6, the millisecond delay adaptive adjustment optimizes the hole millisecond delay, and corrects the reference millisecond delay obtained by optimization based on the distribution characteristics of the rock and mud interlaced interface and the stress wave propagation characteristics.

[0052] The intelligent step blasting system of the rock and mud interlaced open-pit mine is realized based on the intelligent step blasting method of the rock and mud interlaced open-pit mine, and includes:

[0053] A three-dimensional laser scanning module for acquiring point cloud data of the step blasting area of the open-pit mine, including step geometric parameters, rock and mud distribution characteristics and civilian house position;

[0054] A geological modeling and analysis module for processing point cloud data and constructing a three-dimensional geological model reflecting the step shape and rock and mud interlaced distribution characteristics;

[0055] The blasting effect prediction module integrates a blasting vibration prediction model, a flyrock distance prediction model and a fragmentation distribution prediction model, and is used for predicting the vibration response, flyrock risk and fragmentation uniformity under different combinations of blasting parameters.

[0056] The multi-objective optimization module is used for executing a non-dominated sorting genetic algorithm, processing a three-objective optimization function with flyrock control, vibration control and fragmentation uniformity as the targets, and determining the globally optimal blasting parameter benchmark values including the hole pattern parameters, single-hole charge and millisecond delay.

[0057] The parameter adaptive adjustment module is used for adaptively correcting the blasting parameter benchmark values including the hole pattern parameters, single-hole charge and millisecond delay output by the multi-objective optimization module.

[0058] The blasting scheme generation and output module is used for integrating the adaptively adjusted blasting parameters, generating a complete blasting design scheme, and outputting a visual construction guidance file.

[0059] Compared with the prior art, the present application has the following beneficial effects:

[0060] 1. The present application realizes accurate perception of complex geological conditions through three-dimensional laser scanning and rock identification, and provides a reliable basis for differential blasting design.

[0061] 2. The present application establishes a comprehensive blasting prediction model including a blasting vibration prediction model, a flyrock distance prediction model and a Kuz-Ram model, which can comprehensively evaluate the blasting effect and safety, and is suitable for near-distance civilian house protection.

[0062] 3. The multi-objective optimization function constructed by the present application can find a balance among the vibration control target, flyrock control target function and fragmentation uniformity target, and provide a Pareto optimal solution set through an optimization algorithm, thereby realizing global optimization of the blasting parameters.

[0063] 4. The adaptive adjustment mechanism constructed by the present application adjusts and corrects the blasting parameters according to the local rock distribution, realizes the combination of global optimization and local optimization, and improves the uniformity of the blasting effect. BRIEF DESCRIPTION OF DRAWINGS

[0064] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments made with reference to the accompanying drawings:

[0065] Figure 1 is a flowchart of the embodiment of the present application;

[0066] Figure 2 is a system structure schematic diagram of the embodiment of the present application;

[0067] Figure 3 This is a site view of an open-pit mine with alternating rock and mud according to an embodiment of the present invention. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0069] like Figure 1 As shown, the intelligent bench blasting method for open-pit mines with alternating rock and mud includes the following steps:

[0070] Step S1: Obtain point cloud data of the bench blasting area in the open-pit mine through 3D laser scanning;

[0071] Step S2: Construct a three-dimensional geological model that reflects the morphology of the steps and the distribution characteristics of rock and mud;

[0072] Step S3: Based on the characteristics of bench blasting, construct a comprehensive blasting prediction model;

[0073] Step S4: Construct a multi-objective optimization function that includes flystone control, vibration control, and block size uniformity;

[0074] Step S5: The multi-objective optimization function is processed by a multi-objective optimization algorithm to determine the reference hole network parameters, single hole charge, and differential delay for bench blasting. The hole network parameters include hole spacing, row spacing, and over-depth.

[0075] Step S6: Adaptively adjust the reference hole network parameters, single hole charge, and differential delay based on the local rock and mud distribution.

[0076] Step S2 specifically includes the following steps:

[0077] Step S2.1: Perform point cloud registration preprocessing on the acquired point cloud data and use a statistical filtering algorithm to remove noise;

[0078] The filtering function of the statistical filtering algorithm is:

[0079]

[0080] in, This represents the denoised point cloud data. This represents the raw point cloud data. This represents the i-th point in the point cloud. and This represents the mean and standard deviation of a local neighborhood, i.e., removing points in the original point cloud data whose difference from the mean is greater than or equal to three times the standard deviation;

[0081] The point cloud registration preprocessing involves using the iterative nearest point algorithm to accurately register multi-station scan data to a unified coordinate system.

[0082] Step S2.2, extracting step geometry features based on the denoised point cloud data, including step height, slope angle and platform width;

[0083] The step height is calculated by the elevation difference between the step top line and the step bottom line in the point cloud data, i.e. the average elevation of the step top line minus the average elevation of the step bottom line.

[0084] The slope angle is calculated by the angle between the normal vector of the step surface in the point cloud data and the vertical direction, or by the inverse tangent of the height difference and the horizontal distance of the step surface.

[0085] The platform width is calculated by the horizontal distance between the step top line and the step bottom line in the point cloud data, i.e. the projection distance of the step top line and the step bottom line on the horizontal plane.

[0086] Step S2.3, identifying the mud boundary based on color features, curvature features and spatial distribution features;

[0087] Step S2.4, generating a three-dimensional geological model by Delaunay triangulation algorithm, integrating the mud classification results and the location information of the civilian houses.

[0088] The step S2.4 specifically includes: first, projecting the optimized point cloud data onto a two-dimensional horizontal plane to generate a basic triangular mesh; then introducing terrain feature lines as constraint edges to ensure that these important features are accurately expressed in the triangular mesh; finally, mapping the two-dimensional triangular mesh back to the three-dimensional space to restore the three-dimensional terrain surface through the elevation information of the point cloud; integrating the classification results of the mud recognition algorithm into the three-dimensional geological model; assigning each vertex in the triangular mesh with a mud classification attribute, marking the rock region vertices as 1, the mud region vertices as 0, and the transition region vertices as continuous values between 0 and 1 according to the interpolation of adjacent vertices; accurately integrating the location and structure information of the civilian houses into the geological model, obtaining the three-dimensional coordinates of the civilian house corner points through high-precision GPS measurement, and adding these points as fixed points to the point cloud data set during the triangulation process; then inserting constraint edges at the civilian house contour boundary to ensure that the boundary of the triangular mesh in the civilian house area is consistent with the actual situation; finally, assigning special attribute labels to the triangular facets in the civilian house area to highlight the location of the civilian house in the model.

[0089] The step S2.3 specifically includes the following steps:

[0090] Respectively constructing the color feature component, curvature feature component and spatial distribution feature component of each point in the point cloud;

[0091] Integrating the feature components of each point to obtain the mud classification comprehensive feature value of each point, i.e. the probability of the point belonging to rock or mud;

[0092] The calculation formula of the color feature component is:

[0093]

[0094] wherein, represents the color feature component of a point , represents the point index, respectively represent the red channel intensity value, the green channel intensity value and the blue channel intensity value of a point , represents the rock color range set, RGB value (120, 120, 120) to (200, 200, 200), represents the mud color range set, RGB value (80, 60, 40) to (150, 120, 80);

[0095] The formula for calculating the curvature feature component is:

[0096]

[0097] wherein, represents the curvature feature component of a point , represents the gradient operator, represents the unit normal vector at a point , represents the absolute value operator;

[0098] The formula for calculating the spatial distribution feature component is:

[0099]

[0100] wherein, represents the spatial distribution feature component of a point , represents the total number of points in the neighborhood of a point , represents the jth point in the neighborhood of a point , represents the indicator function, taking 1 if the condition is true, otherwise taking 0, represents the point set belonging to the same material region as a point .

[0101] The formula for calculating the rock-mud classification comprehensive feature value is:

[0102]

[0103] wherein, represents the rock-mud classification comprehensive feature value of a point , , and These represent the weights of the color feature component, curvature feature component, and spatial distribution feature component, respectively.

[0104] For areas where rock and mud intermingle, the weights are set as follows: =0.33, =0.33, =0.34, when the color contrast between the rock and mud is strong, it increases. The value increases when the surface of the rock and mud has obvious unevenness. The value increases when the mudstone is distributed in continuous layers. value.

[0105] In step S3, the comprehensive blasting prediction model includes three models: a blasting vibration prediction model that considers elevation effects and rock mass damage; a fly rock distance prediction model that integrates surface fly rocks, fly rocks in the blocking section, and fly rocks from geological defects; and a Kuz-Ram model oriented towards block size prediction.

[0106] The calculation formula for the blasting vibration prediction model is as follows:

[0107]

[0108] in, Indicates the peak vibration velocity of a particle. Indicates the dosage per orifice. Indicates the distance between the centers of the explosion. This indicates the elevation difference between the explosion center and the measuring point. Indicates the relative height of the measuring points inside the steps. Indicates the height of the step. Indicates rock mass damage factor. This represents the site foundation coefficient, reflecting the comprehensive impact of site geological conditions on vibration propagation, with a value range of 50-300. This represents the elevation influence coefficient, reflecting the amplification effect of elevation difference on vibration, with a value range of 0.02-0.08. This represents the damage attenuation coefficient, reflecting the degree to which rock mass damage promotes vibration attenuation, and its value ranges from 0.1 to 0.3. This represents the influence coefficient of step height, reflecting the amplification effect of the height of the measuring point inside the step on vibration. Its value ranges from 0.1 to 0.3. This represents the dosage index, reflecting the degree of nonlinear influence of dosage on vibration intensity, with a value ranging from 0.5 to 0.8. This represents the distance index, reflecting the degree of influence of distance on vibration attenuation, with a value range of 1.2-1.8.

[0109] Rock mass damage factor It is obtained by subtracting the difference in rock quality index RQD from 100 and then dividing by 100;

[0110] The specific formula of the flyrock distance prediction model is:

[0111]

[0112] Wherein, represents the flyrock distance prediction model, represents the surface flyrock distance, which is calculated based on the blast hole diameter and explosive parameters, represents the plug section flyrock distance, which is calculated based on the plug length and mass, represents the geological defect flyrock distance, which is calculated based on the joint density and width;

[0113] Wherein, the specific calculation formula of the surface flyrock distance, the plug section flyrock distance and the geological defect flyrock distance is:

[0114]

[0115]

[0116]

[0117] Wherein, represents the blast hole diameter, represents the explosive density, represents the actual plug length, represents the required plug length, represents the joint density, represents the joint width;

[0118] The Kuz-Ram model oriented to block degree prediction is composed of Kuznetsov equation, R-R distribution function and uniformity index, and the calculation formula is as follows:

[0119]

[0120]

[0121]

[0122] Wherein, is the average broken block degree; is the rock coefficient, and Kuznetsov believes that the range of rock coefficient is 7-13; is the unit explosive consumption; is the single-hole charge; is the relative mass power of explosive, which is 100 for ammonium oil explosive and 110 for emulsified explosive; is the uniformity index; is the minimum resistance line; is the blast hole diameter; is the hole spacing; is the drilling accuracy standard deviation, generally taken as 0.05H; is the bench height; is the length of the charge above the floor level; is the percentage of rock mass less than a certain particle size; is the screen size, is the characteristic size of the rock mass, that is, the size of the block size when the cumulative undersize is about 63.21%.

[0123] In the step S4, the specific formula of the multi-objective optimization function is:

[0124]

[0125] wherein, is a multi-objective optimization function vector composed of three objective function values, is a vibration control objective function, is a flyrock control objective function, is a block size uniformity objective function, is a blasting parameter vector.

[0126] wherein, the specific formula of the vibration control objective function, the flyrock control objective function and the block size uniformity objective function is:

[0127]

[0128]

[0129]

[0130] wherein, is the maximum vibration velocity calculated by the blasting vibration prediction model under the blasting parameter vector , is the maximum allowable vibration velocity, taken as 2.5 cm / s, is the minimum flyrock distance calculated by the flyrock distance prediction model under the blasting parameter vector , is the safety distance, taken as 50 m, is the block size distribution standard deviation calculated by the Kuz-Ram model under the blasting parameter vector , is the average broken block size calculated by the Kuz-Ram model under the blasting parameter vector .

[0131] The constraint conditions of the multi-objective optimization function include three levels of safety constraints, technical feasibility constraints and effect guarantee constraints.

[0132] The safety constraints mainly include blasting vibration safety constraints and flying rock safety constraints. The blasting vibration safety constraints require that the predicted peak particle vibration velocity should not exceed the allowed safety threshold to ensure the safety of residential structures. The flying rock safety constraints require that the predicted maximum flying distance of flying rocks should be less than the safety distance and leave enough safety margin to prevent flying rocks from causing damage to residential structures.

[0133] The technical feasibility constraints involve the range limits of blasting parameters, including hole spacing, row spacing, overburden, single-hole charge and upper and lower limit constraints of millisecond delay. These constraints ensure that the blasting parameters are within the feasible range of engineering practice, while the stemming length must meet the minimum requirement to ensure blasting safety and energy utilization rate.

[0134] The effect guarantee constraints are mainly aimed at the mass of blasting block size, which requires that the uniformity index of block size distribution should not exceed the allowed limit to ensure uniform block size after blasting, facilitate subsequent mining and loading operations and reduce secondary crushing costs.

[0135] All constraint conditions are treated as hard constraints in the optimization process, and any solution that violates the constraint conditions will be directly eliminated.

[0136] In the step S5, the multi-objective optimization algorithm adopts a non-dominated sorting genetic algorithm, including the following specific steps:

[0137] Step S5.1, randomly generate an initial population, each individual in the population is a solution vector containing hole network parameters, single-hole charge and millisecond delay;

[0138] Step S5.2, for each individual in the population, calculate the values of the three objective functions in the multi-objective optimization function respectively;

[0139] Step S5.3, non-dominant sorting of the population according to the objective function values;

[0140] Step S5.4, calculate the crowding degree of each solution;

[0141] Step S5.5, use binary tournament selection to select parent individuals, then perform crossover and mutation operations on the selected parent individuals to generate a child population;

[0142] Step S5.6, combine the parent population and the child population into a new population;

[0143] Step S5.7, non-dominant sorting and crowding degree calculation of the new population again, and finally select the next generation population;

[0144] Step S5.8, stopping outputting the Pareto optimal solution set when the maximum iteration number 200 is reached;

[0145] Step S5.9, selecting an optimal solution, i.e., a reference hole pattern parameter, a single-hole charge weight and a millisecond delay, from the Pareto optimal solution set.

[0146] The parameter initialization range is specifically: hole spacing 3.0-5.5m, row spacing 2.5-4.2m, super depth 0.2-1.8m, single-hole charge weight 30-180kg, and millisecond delay 15-150ms.

[0147] In the step S6, the hole pattern parameter adaptive adjustment specifically includes hole spacing adaptive adjustment, row spacing adaptive adjustment and super depth adaptive adjustment.

[0148] The hole spacing adaptive adjustment comprehensively considers the reference hole spacing, the local rock and mud distribution proportion and the lithology adjustment coefficient, and scales the reference hole spacing according to the local rock rate and the mud rate.

[0149] The row spacing adaptive adjustment establishes an association relationship with the hole spacing through a row spacing coefficient based on the adjusted hole spacing and the step height factor, and considers the correction when the step height deviates from the standard value.

[0150] The super depth adaptive adjustment comprehensively considers the reference super depth, the local lithology difference and the step height influence, adjusts the super depth value according to the proportion difference between rock and mud, and compensates for the non-standard step height.

[0151] The formula of the hole pattern parameter adaptive adjustment is:

[0152]

[0153]

[0154]

[0155] wherein, represents the adaptive adjusted hole spacing, represents the optimized reference hole spacing, represents the local rock rate, i.e., the proportion of rock within 5m around the current blast hole, represents the local mud rate, i.e., the proportion of mud within 5m around the current blast hole, represents the adaptive adjusted row spacing, represents the adaptive adjusted super depth, represents the optimized reference super depth.

[0156] In the step S6, the single-hole charge amount self-adaptive adjustment is based on the benchmark single-hole charge amount obtained by optimization, and dynamically adjusted in combination with local lithological characteristics, relative distance between the blast hole and the civilian house, and vibration control requirements.

[0157] The formula of the single-hole charge amount self-adaptive adjustment is:

[0158]

[0159] wherein, represents the single-hole charge amount after self-adaptive adjustment, represents the benchmark single-hole charge amount obtained by optimization, represents a distance attenuation factor, and takes a value of 0.7-0.8, represents a vibration control factor, and takes a value in a range of 0.7-1.0.

[0160] In the step S6, the millisecond delay self-adaptive adjustment is performed on the millisecond delay in the hole, and the benchmark millisecond delay obtained by optimization is corrected based on the distribution characteristics of the rock-mud interbedded interface and the stress wave propagation characteristics.

[0161] The formula of the millisecond delay self-adaptive adjustment is:

[0162]

[0163] wherein, represents the millisecond delay after self-adaptive adjustment, represents the benchmark millisecond delay obtained by optimization, represents a rock-mud interface additional delay, and takes a value in a range of 10-25 ms.

[0164] As shown in Figure 2 the intelligent bench blasting system for the rock-mud interbedded open-pit mine includes:

[0165] a three-dimensional laser scanning module, configured to acquire point cloud data of a bench blasting area of the open-pit mine, including bench geometric parameters, rock-mud distribution characteristics, and a civilian house position;

[0166] a geological modeling and analysis module, configured to process the point cloud data, and construct a three-dimensional geological model reflecting the bench shape and the rock-mud interbedded distribution characteristics;

[0167] a blasting effect prediction module, integrated with a blasting vibration prediction model, a flyrock distance prediction model, and a fragmentation distribution prediction model, and configured to predict vibration responses, flyrock risks, and fragmentation uniformity under different combinations of blasting parameters;

[0168] A multi-objective optimization module is configured to execute a non-dominated sorting genetic algorithm to process a three-objective optimization function targeting flyrock control, vibration control and block uniformity, and determine globally optimal benchmark values of blasting parameters including hole pattern parameters, single-hole charge and millisecond delay;

[0169] A parameter self-adaptive adjustment module is configured to perform self-adaptive correction on the benchmark values of blasting parameters including hole pattern parameters, single-hole charge and millisecond delay output by the multi-objective optimization module;

[0170] A blasting scheme generation and output module is configured to integrate the self-adaptively adjusted blasting parameters, generate a complete blasting design scheme, and output a visualized construction guidance file.

[0171] As shown in Figure 3 , it is a construction site after open-pit bauxite bench blasting, and the soil and rock are interlaced. The difficulty of blasting at this place mainly lies in that the rock is divided by mud and is very close to a residential house, about 50 m. Both blasting effect and vibration reduction should be ensured to guarantee safety.

[0172] The examples described in the present application are merely to describe the preferred embodiments of the present application, and are not intended to limit the concept and scope of the present application. Without departing from the design idea of the present application, various modifications and improvements to the technical solutions of the present application made by the engineering and technical personnel in the field shall fall within the protection scope of the present application.

Claims

1. An intelligent bench blasting method for a rock soil interlaced open-pit mine, characterized in that, The method comprises the following steps: Step S1, obtaining bench blasting area point cloud data of an open-pit mine by three-dimensional laser scanning; Step S2, constructing a three-dimensional geological model reflecting the bench shape and the distribution characteristics of the rock and mud; Step S3, constructing a comprehensive blasting prediction model based on the characteristics of bench blasting; Step S4, constructing a multi-objective optimization function including fly rock control, vibration control and block uniformity; Step S5, determining the bench blasting reference hole pattern parameters, single-hole charge and millisecond delay by processing the multi-objective optimization function through a multi-objective optimization algorithm, wherein the hole pattern parameters include hole spacing, row spacing and overburden depth; Step S6, adaptively adjusting the reference hole pattern parameters, single-hole charge and millisecond delay according to the local rock and mud distribution.

2. The method of claim 1, wherein, The step S2 specifically comprises the following steps: Step S2.1, performing point cloud registration preprocessing on the obtained point cloud data, and removing noise by using a statistical filtering algorithm; Step S2.2, extracting bench geometric features including bench height, slope angle and platform width based on the denoised point cloud data; Step S2.3, identifying the rock and mud boundary based on color features, curvature features and spatial distribution features; Step S2.4, generating a three-dimensional geological model by using a Delaunay triangulation algorithm, and integrating the rock and mud classification results and the location information of the houses.

3. The method of claim 2, wherein, The step S2.3 specifically comprises the following steps: Respectively constructing a color feature component, a curvature feature component and a spatial distribution feature component of each point in the point cloud; Obtaining a rock and mud classification comprehensive feature value of each point by comprehensively integrating the feature components of each point, i.e. the probability that the point belongs to rock or mud; The calculation formula of the color feature component is as follows: ； wherein, a color feature component of a point, a point index, respectively represent a red channel intensity value, a green channel intensity value and a blue channel intensity value of a point, represent a rock color range set, RGB values (120, 120, 120) to (200, 200, 200), represent a mud color range set, RGB values (80, 60, 40) to (150, 120, 80);​​ The calculation formula of the curvature feature component is as follows: ； wherein denotes a curvature feature component of a point , denotes a gradient operator, denotes a unit normal vector at a point , denotes an absolute value operator; The calculation formula of the spatial distribution feature component is as follows: ； wherein, a spatial distribution feature component of the point , a total number of points in the neighborhood of the point , a j-th point in the neighborhood of the point , denotes an indicator function, which takes 1 if the condition is true, and 0 otherwise, denotes a set of points belonging to the same material region as the point .

4. The method of claim 3, wherein, In the step S3, the comprehensive blasting prediction model comprises three models, which are respectively: a blasting vibration prediction model considering the elevation effect and rock mass damage; a fly rock distance prediction model comprehensively considering surface fly rock, jammed section fly rock and geological defect fly rock; and a Kuz-Ram model for block size prediction.

5. The method of claim 4, wherein, In the step S4, the specific formula of the multi-objective optimization function is as follows: ； wherein, represents a multi-objective optimization function vector composed of three objective function values, represents a vibration control objective function, represents a flyrock control objective function, represents a lumpiness uniformity objective function, represents a blasting parameter vector.

6. The method of claim 5, wherein, In the step S5, the multi-objective optimization algorithm uses a non-dominated sorting genetic algorithm, which comprises the following specific steps: Step S5.1, randomly generating an initial population, each individual in the population being a solution vector containing hole pattern parameters, single-hole charge and millisecond delay; Step S5.2, for each individual in the population, calculating the values of the three objective functions in the multi-objective optimization function respectively; Step S5.3, non-dominantly sorting the population according to the objective function values; Step S5.4, calculating the crowding degree of each solution; Step S5.5, using binary tournament selection to select parent individuals, and then performing crossover and mutation operations on the selected parent individuals to generate a child population; Step S5.6, merging the parent population and the child population into a new population; Step S5.7, non-dominantly sorting and calculating the crowding degree of the new population again, and finally selecting the next generation population; Step S5.8, stopping outputting the Pareto optimal solution set when the maximum number of iterations 200 is reached; Step S5.9, selecting an optimal solution from the Pareto optimal solution set, i.e., the benchmark hole pattern parameters, single-hole charge and millisecond delay.

7. The method of claim 6, wherein, In the step S6, the hole pattern parameter adaptive adjustment specifically includes hole spacing adaptive adjustment, row spacing adaptive adjustment and super-deep adaptive adjustment; The hole spacing adaptive adjustment comprehensively considers the benchmark hole spacing, local rock and mud distribution proportion and lithology adjustment coefficient, and scales the benchmark hole spacing according to the local rock rate and mud rate; The row spacing adaptive adjustment establishes an association relationship with the hole spacing through a row spacing coefficient based on the adjusted hole spacing and step height factor, and considers the correction when the step height deviates from the standard value; The super-deep adaptive adjustment comprehensively considers the benchmark super-deep, local lithology difference and step height influence, adjusts the super-deep value according to the proportion difference between rock and mud, and compensates for the non-standard step height.

8. The method of claim 7, wherein, In the step S6, the single-hole charge adaptive adjustment is based on the benchmark single-hole charge obtained by optimization, and dynamically adjusts the single-hole charge in combination with the local lithology characteristics, the relative distance between the blast hole and the civilian house and the vibration control requirement.

9. The method of claim 8, wherein, The millisecond delay adaptive adjustment optimizes the hole millisecond delay, and corrects the benchmark millisecond delay obtained by optimization based on the distribution characteristics of the rock and mud interbedded interface and the stress wave propagation characteristics.

10. An intelligent bench blasting system for a rock soil interlaced open pit mine, which is implemented based on the intelligent bench blasting method for a rock soil interlaced open pit mine according to any one of claims 1-9, characterized in that, It comprises: A three-dimensional laser scanning module for obtaining point cloud data of the bench blast area of the open-pit mine, including bench geometric parameters, rock and mud distribution characteristics and civilian house position; A geological modeling and analysis module for processing point cloud data and constructing a three-dimensional geological model reflecting the bench shape and rock and mud interbedded distribution characteristics; A blasting effect prediction module integrating a blasting vibration prediction model, a flyrock distance prediction model and a fragmentation distribution prediction model for predicting the vibration response, flyrock risk and fragmentation uniformity under different blasting parameter combinations; A multi-objective optimization module for executing a non-dominated sorting genetic algorithm to process a three-objective optimization function with flyrock control, vibration control and fragmentation uniformity as the objectives, and determine the globally optimal benchmark values of the blasting parameters including hole pattern parameters, single-hole charge and millisecond delay; A parameter adaptive adjustment module for adaptively correcting the blasting parameter benchmark values including hole pattern parameters, single-hole charge and millisecond delay output by the multi-objective optimization module; A blasting scheme generation and output module for integrating the adaptively adjusted blasting parameters, generating a complete blasting design scheme and outputting a visual construction guidance file.

Images